Occurrence of Aflatoxin M₁ in Vacuum Packed Kashar Cheeses in Turkey

Abstract

In this study, aflatoxin M₁ (AFM₁) levels of vacuum packed Kashar cheese marketed in the Black Sea Region of Turkey were determined during October 2008 and June 2009. The occurrence of AFM₁ contamination in vacuum packed cheese samples was investigated using immunoaffinity column extraction and high-performance liquid chromatography. AFM₁ was detected in 144 (97.96%) of cheese samples ranging from 0.015 to 3.774 μg/kg cheese, and mean value of AFM₁ was found to be 0.273 μg/kg cheese. AFM₁ level in 16 (11.11%) of positive samples was higher than the maximum tolerance limit (0.5 μg/kg) accepted by Turkish Food Codex.

Keywords:

• Aflatoxin M₁
• Kashar cheese
• HPLC
• Immunoaffinity column
• Black Sea

INTRODUCTION

Mycotoxins are a group of secondary metabolites of various filamentous fungi species that are the result of fungal growth in foodstuff, crops, and human foods.[1] Most of the mycotoxins are toxic to human beings and animals, causing several chronic and acute health problems, such as kidney damage, cancer, and immune suppression.[2] Aflatoxins are a major group of mycotoxins and are mainly produced by several Aspergillus species, including Aspergillus flavus, A. parasiticus, A. nominus, A. pseudotomarii, A. bombycis, and A. tamari, in addition to various species of Penicillium, Rhizopus, Mucor, and Streptomyces that contaminated to plant and plant products.[3,4] The major aflatoxins of concern are called B₁, B₂, G₁, and G₂. These toxins are usually found together in foods and animal feeds with different ratios.[5] A. flavus produces only aflatoxin B₁ (AFB₁), the most potent carcinogen, and aflatoxin B₂ (AFB₂), while the other species may produce all forms of aflatoxin.[6] Aflatoxin M₁ (AFM₁) and M₂ (AFM₂), the main monohydroxilated metabolically transformed products of AFB₁ and AFB₂, may be found in milk and dairy products, such as cheese, dried milk, and yoghurt.[7] There is a positive relationship between the amount of AFM₁ in milk and AFB₁ in the feed consumed by the animals.[8,9] Different percentages of AFB₁, ranging between 0.3 and 6.2%, in animal feed have been reported to be transformed to AFM₁ and excreted in milk.10–13 13] Transformation rate depends on numerous factors, including the animal species, the stage of lactation, the moment of milking, the variability of individuals, the productive level, and udder health.[4]

It has been well demonstrated that AFM₁ provided a source of potential health problems causing toxic and carcinogenic, and also genotoxic and mutagenic defects.[14,15] The Internal Agency for Research on Cancer (IARC) has classified AFM₁ as a Group 2 with respect to human carcinogenicity, while it was transferred to Group 1 in 2002.16–18 18] Therefore, many countries have improved their research and control programs to reduce the potential risk of AFM₁ and revised the maximum acceptable limits of aflatoxins in feeds and foodstuffs.[19] According to the Turkish Food Codex, the maximum permissible limit of AFM₁ in cheese is 0.5 μg/kg.[20]

Kashar, a traditional semi-hard cheese, is one of the most popular cheeses manufactured in Turkey.[21] According to Turkish Standards, Kashar cheese is classified as “fresh Kashar cheese” and “old or ripened Kashar cheese” depending on the ripening period.[22] In recent years, the production of fresh (vacuum) Kashar has increased in contrast to ripened Kashar cheese because of economical reasons.[23] The Black Sea Region, in the northern part of Turkey, sees the most rain in Turkey, with average annual rainfall of over 1.000 mm, and the temperature changes moderately during the year from −2 to 30°C, with a humidity of 78%.[24] Thereby, suitable conditions for growth of mycotoxigenic molds like A. flavus and A. parasiticus in foodstuffs are available in this region. Although several studies have been carried out to investigate fungal contamination and mycotoxin levels of Kashar cheese and dairy product in various regions of Turkey,[5,25–30 30] to the best of our knowledge, there is no study about AFM₁ levels of Kashar cheeses manufactured and marketed in the Black Sea Region in Turkey. The present study was conducted to determine the natural occurrence and levels of AFM₁ in vacuum packed Kashar cheeses produced and marketed in the Black Sea Region of Turkey in different seasons and to compare the results with the maximum AFM₁ tolerance limit accepted by the Turkish Food Codex.

MATERIALS AND METHODS

Materials

In our study, a total of 147 samples of vacuum packed Kashar cheese were obtained randomly from markets, manufactured by 49 different dairy plants during the periods of October–November (1st period), February–March (2nd period), and May–June (3rd period) in the East (the cities of Amasya, Bayburt, Giresun, Gümüşhane, and Trabzon), Middle (the cities of Çorum, Ordu, Samsun, Sinop, and Tokat), and West (the cities of Bartın, Bolu, Düzce, Karabük, Kastamonu, and Zonguldak) Black Sea Region, in 2008–2009 (Fig. 1). Cheese samples were taken from 0.5 to 1 kg quantities and transported to the laboratory in an insulated container at about 4°C for analysis.

Figure 1 Map of Turkey with the locations of sampling sites. (Color figure available online.)

Chemicals

High-performance liquid chromatography (HPLC) grade solvents (acetonitrile, hexane, methanol, and chloroform) were purchased from Merck (Darmstadt, Germany). AFM₁ standard was obtained from Supelco (Bellefonte, PA, USA) and afla M₁ immunoaffinity columns were obtained from VICAM (Watertown, MA, USA).

Methods

Sample preparation

Cheese samples were prepared according to Sharman et al.[31] First, 20 g of minced cheese, 5 g of Celite-545, 75 mL of chloroform, and 0.5 g of saturated sodium chloride were blended using an Ultra Turrax (Daihan Scientifics, WiseMix, HG-15D, Seoul, Korea) at medium speed for 3 min to form slurry. The slurry was filtered through a Whatman no. 4 filter paper into a round-bottomed flask. The beaker was washed with 50 mL of chloroform and the washings were filtered. The chloroform extract was evaporated to dryness under vacuum at 30°C using a laboratory type rotary evaporator (Buchi Rotavapor R-200, Flawil, Switzerland). One milliliter of methanol, 30 mL of water, and 50 mL of hexane were added to the residue and the mixture was shaken. The mixture was transferred to a separating funnel with washing (2 × 10 mL of water) and shaken for 10–15 s. The layers were allowed to separate and the lower layer was collected and used in the affinity column stage of the sample clean-up.

Immunoaffinity column clean-up and elution

Sample extract was collected from above the separating funnel and passed from the column at a flow rate of about 1–2 mL/min. Then the column was washed with distilled water (10 mL) twice and air was passed from the column using a syringe. AFM₁ was eluted with acetonitrile (1 mL) and distilled water (1 mL). Elute was homogenized using a vortex and 100 μL of elute was injected to HPLC.

Determination of AFM₁ by HPLC

AFM₁ analysis was done by a HPLC system (Shimadzu, Kyoto, Japan) consisting of a pump (Shimadzu LC-20AT), fluorescence detector (Shimadzu RF-10XL), excitation and emission wavelengths were 365 and 435 nm, respectively, C18 column (5 μm, 250 mm, inertsil ODS-3, GL Sciences Inc., Tokyo, Japan), column oven (Shimadzu CTO-10AS VP) set to 23°C, and auto sampler (Shimadzu SIL-10A). Mobil phase was acetonitrile-water (25:75 v/v) at 1 mL/min. Injection volume was assayed as 100 μL.

Preparation of AFM₁ standard

Stock solution of AFM₁ was prepared according to the method described by AOAC.[32] First, stock solution (buffer stock) was diluted with acetonitrile to 2 mL. Then, the last concentrations (0.05, 0.1, 0.2, 0.5, and 5 μg/L) for calibration curve were prepared from buffer stock and were kept at 4°C.

Statistical Analysis

Statistical analysis was performed using SPSS version 12.0 (SPSS Inc., Chicago, IL, USA). ANOVA was used to determine the effect of region and period and Duncan test was applied for multiple comparison (α = 0.05).

RESULTS AND DISCUSSION

The standard curve was obtained for AFM₁ concentration between 0.05–5 μg/L (R ² = 0.9999). The chromatograms of AFM₁ standard at 5 μg/L (a), AFM₁ added cheese samples at 1 μg/L (b), and analyzed cheese samples at 3.774 μg/kg (c) were shown in Fig. 2. The detection limit was 0.01 μg/kg. Samples with AFM₁ levels below 0.01 μg/kg were considered as negative. Recovery of AFM₁ added to cheese at 0.25 μg/kg averaged 88.3% and analytical results were corrected for recovery.

Figure 2 (a) HPLC chromatograms of AFM₁ standard at 5 μg/L, (b) cheese samples with AFM₁ at 0.5 μg/L, and (c) analyzed cheese samples at 3.774 μg/kg. (Color figure available online.)

The occurrence and distribution of AFM₁ in vacuum Kashar cheese samples were presented in Tables 1 and 2. As shown in Table 1, AFM₁ was detected in 144 samples of 147 total cheese samples at a mean concentration of 0.273 μg/kg and at the levels ranging between 0.015–3.774 μg/kg. AFM₁ levels in 16 (11.11%) of positive samples were found to be higher than the maximum acceptable limits (0.5 μg/kg for cheese, Table 2). The highest mean concentration of AFM₁ was determined in February–March samples (0.431 μg/kg) and the lowest in May–June samples (0.166 μg/kg). Furthermore, cheese samples collected in West and East Black Sea Regions exhibited the highest and lowest mean concentrations of AFM₁ (0.321 and 0.206 μg/kg), respectively (Table 1).

Table 1 AFM₁ levels of vacuum packed Kashar cheeses produced in different periods (μg/kg)

Region	Period	N	Min.	Max.	X ^a	Sd ^b
East Black Sea	1	13	0.019	0.408	0.145	0.112
	2	13	0.023	0.491	0.275	0.16
	3	13	0.042	0.426	0.196	0.13
Middle Black Sea	1	13	0.021	0.983	0.257	0,333
	2	13	0.074	1.508	0.512	0.532
	3	13	ND	0.215	0.106	0.066
West Black Sea	1	23	0.015	1.2	0.26	0.282
	2	23	0.053	3.774	0.506	0.833
	3	23	ND	1.368	0.196	0.294
Black Sea	1	49	0.015	1.2	0.221	0.242
	2	49	0.023	3.774	0.431	0.508
	3	49	ND	1.368	0.166	0.163
Overall samples		147	0.015	3.774	0.273	0.305
^aMean.
^bStandard deviation.
ND: Not detected; N: number of the samples analyzed.

Table 2 Distribution of AFM₁ levels in vacuum packed Kashar cheeses produced in East, Middle, and West Black Sea Regions

	AFM1 levels of Kashar cheeses (μg/kg) and percentage
Region		ND–0.05	0.05–0.1	0.1–0.25	0.25–0.5	0.5–1	1<
East Black Sea	ND ^a	4	9	14	12	0	0
	% ^b	10.26	23.08	35.9	30.77	0	0
Middle Black Sea	ND ^a	7	11	9	5	4	3
	% ^b	17.95	28.21	23.08	12.82	10.26	7.69
West Black Sea	ND ^a	12	7	23	18	5	4
	% ^b	17.39	10.14	33.33	26.09	7.25	5.8
Total	ND ^a	23	27	46	35	9	7
	% ^b	15.65	18.37	31.29	23.81	6.12	4.76
^aCheese samples.
^bPercentage in total samples.
ND: Not detected.

The incidence of AFM₁ contamination in cheese was found to be high, since 97.96% of the samples were positive (Table 2). This result was in accordance with previous studies performed in Turkey. Tekinsen and Eken[33] analyzed 132 Kashar cheese samples and found 82.6% of them contaminated with AFM₁. The number of cheese samples exceeding the acceptable limits of 0.25 μg/kg (old legal limit according Turkish Food Codex) were 36 (27.27%). Tekinsen and Ucar[9] examined AFM₁ levels of 100 cream cheese samples collected from retail outlets in five big cities (Istanbul, Izmir, Kayseri, Konya, and Tekirdag) and found AFM₁ in 99% of the samples and 18% of them had AFM₁ levels exceeding 0.25 μg/kg. Ardic et al.[34] showed that 82.4% of 192 white cheese samples were contaminated with AFM₁, and AFM₁ contamination above 0.25 μg/kg was found in 51 of the cheese samples. AFM₁ was detected in 85.46%,[26] 95%,[28] and 91.81%[35] of cheese samples. The results of Alborzi et al.,[36] Elbergi et al.,[37] and Prado et al.[38] were similar to the present study. On the other hand, results of some studies related with AFM₁ in cheese in Turkey[12,39,40] and other countries[4,41,42] were lower than our results. These differences may be due to the different analysis methods, cheese type and process, softness or hardness of cheese, applied heat treatments, and proteolysis.[5,10] Moreover, differences in the hygiene and storage conditions at the dairies and retail points are other key factors on the variations of the results.[33]

The mean levels of AFM₁ were found to be 0.206, 0.292, and 0.321 μg/kg in the samples of East, Middle, and West Black Sea Regions, respectively (Fig. 3) and no statistical difference (P > 0.05, Table 3) was present between them. Even if there were no statistical differences between the mean AFM₁ levels of the Kashar cheese samples in the different regions, the cheeses produced in the East Black Sea Region could be considered as safer for human consumption than other cheeses due to the low mean concentration of AFM₁ level and the number of cheese samples that exceed the legal limits. For this reason, the feeding practices are varied in these regions. In the East Black Sea Region, there is an abundance of green fodder and it is used as a feed for cattle. Another factor for less concentration of AFM₁ in milk samples from this region is a common practice of grazing. In the Middle and West Black Sea Regions, there is less availability of green fodder and there is excessive use of concentrated feed, such as silage.

Table 3 The effect of season and region on AFM₁ levels of vacuum packed Kashar cheese produced in Black Sea Region

Period	N	Mean of AFM1 concentration (μg/kg)	Region	N	Mean of AFM1 concentration (μg/kg)
First	49	0.221^a	East Black Sea	39	0.206^a
Second	49	0.431^b	Middle Black Sea	39	0.292^a
Third	49	0.166^a	West Black Sea	69	0.321^a
Means followed by the same letter are significantly different (P > 0.05) according to Duncan's multiple range test. N: Number of the samples analyzed.

Figure 3 AFM₁ concentrations of vacuum packed Kashar cheeses produced in different periods in East, Middle, and West Black Sea Region (μg/kg).

The mean AFM₁ amounts of the cheeses collected in the first (October–November), second (February–March), and third periods (May–June) were 0.221, 0.431, and 0.166 μg/kg, respectively (Fig. 3). Statistical evaluation showed that there were significant differences (P < 0.05) between the mean concentrations of AFM₁ levels of cheese samples collected in different periods (Table 3). In the second period, the mean AFM₁ level was the highest (P < 0.05), but at first and third periods, the means of AFM₁ were similar (Table 3). This may result from high levels of AFB₁ of the feeds eaten by the lactiferous animals in the second period (February–March), since it was reported that there was a relationship between AFM₁ occurrence level in milk and AFB₁ content of the feed. Additionally, milk producers may provide animal feeds with low cost in autumn and store them with silage method. Thereby, a favorable media may occur for growth of mycotoxigenic molds like A. flavus and A. parasiticus leading to the increase of AFB₁ concentration in feeds.[43] It is similar to the reports of several authors, which emphasized a higher incidence of contamination during cold seasons rather than hot ones. In one study, Torkar and Vengust[44] detected higher AFM₁ concentrations in the raw milk samples produced in cold seasons than the levels of the products of hot seasons. Ghiasian et al.[45] collected 186 milk samples in summer and winter seasons and showed that the presence of AFM₁ was 56.6 and 71.8% in those seasons, respectively. Bakirci[8] reported that AFM₁ levels in June milks were lower than March–April–May milks. Galvano et al.[46] informed that AFM₁ levels of collected milk samples in October–April were four times higher than May–September ones. Overall, the results of the present study were in accordance with the results of previous studies. However, Markaki and Melissari[47] determined that the season had no effect on AFM₁ levels in milk products.

CONCLUSION

The presence of high levels of AFM₁ in Kashar cheese in Turkey seems to still be a main concern for public health. This study indicated that the AFM₁ concentration of the cheeses showed a seasonal variation and varied in a wide range depending on the region where the samples were obtained. Thus, it may also be due to poor storage conditions of the products or contamination of animal feedstuffs with high levels of AFB₁, which is further transmitted to the milk and dairy products after its conversion to AFM₁ in the animal metabolism. The following suggestions should be taken into account to produce safe Kashar cheese with low levels of AFM₁: (i) feedstuffs, which are consumed by animals, should be prevented from the contamination of molds and AFB₁ during the harvest and storage periods, (ii) dairy farmers should be educated by the government authorities about the potential health consequences of aflatoxins, and (iii) AFB₁ levels in feeds and AFM₁ levels in dairy products should be monitored and controlled periodically.

ACKNOWLEDGMENT

This study was financially supported by Ondokuz Mayis University Research Foundation (MF-136).